Bifidobacterium longum as a delivery system of TRAIL and endostatin cooperates with chemotherapeutic drugs to inhibit hypoxic tumor growth

1

Revision of CGT-08-0083R

Bifidobacterium Longum as a delivery system of TRAIL and

endostatin cooperates with chemotherapeutic drugs to inhibit

hypoxic tumor growth

Bi Hu1, Lei Kou1, Chen Li1, Li-Ping Zhu1, Yan-Rong Fan2, Zhi-Wei Wu3, Jian-Jun Wang1,*, Gen-Xing Xu3,4,*

1 Department of Biological Science and Technology and State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Mailbox 426, Nanjing

University, 22 Hankou Road, Nanjing 210093, China 2School of Chemical Engineering, Nanjing University of Science and Technology,

Nanjing 210094, China 3Center for Public Health Research, Medical School, Nanjing University, 22 Hankou

Road, Nanjing 210093, China 4Jiangsu Provincial Research Center for Gene Pharmaceutical Engineering and

Technology, 42 Chenghu Road, Suzhou 215128, China

Running Head: TRAIL in B. longum inhibit tumor growth

*Corresponding authors:

Gen-Xing Xu Jian-Jun Wang Center for Public Health Research School of Life Sciences, Mailbox 426 Medical School, Nanjing University Nanjing University 22 Hankou Road 22 Hankou Road Nanjing 210093 Nanjing 210093 China China Tel & Fax: +86-25-83597570 Tel & Fax: +86-25-83592714 Email: [email protected] Email: [email protected]

2

Abstract

In our previous study, we have shown that vector pBV22210 containing a

chloramphenicol resistance and a cryptic plasmid pMB1 from Bifidobacterium

longum strain could stably replicate and did not significantly affect the biological

characteristics of Bifidobacterium longum (B. longum). In the current study, B.

longum was transfected by electroporation with pBV22210 encoding the extracellular

domain of TRAIL (B. longum-pBV22210-TRAIL) and its carbohydrate fermentation

and growth curve were determined, and its location and inhibitory effect on tumor

xenografts in mice were also examined. The results further proved that gene

transfection did not change the main biochemical characteristics of Bifidobacterium

longum. The results also showed that B. longum-pBV22210-TRAIL resulted in

selective location in tumors and exhibited a definite antitumor effect on S180

osteosarcoma. In addition, when a low dosage of Adriamycin (5 mg/kg) or B.

longum-pBV22210-endostatin was combined, the antitumor effect was significantly

enhanced. The successful inhibition of S180 tumor growth suggested a stable vector

in Bifidobacterium longum for transporting anticancer genes combined with low dose

chemotherapeutic drugs or other target genes is a promising approach in cancer gene

therapy.

Key words: Bifidobacterium longum, TRAIL, B. longum-pBV22210-TRAIL,

endostatin, gene therapy, selective location, synergistic interactions

3

Introduction

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is a member of

tumor necrosis factor superfamily.1,2 TRAIL triggers apoptosis through binding to its

membrane receptors (DR4 and DR5) and activating caspase-8 in majority of tumor

cells but not in normal cells.3-8 Recombinant soluble human TRAIL is thought to be a

candidate for cancer therapy because of its potent antitumor effects without toxicity in

normal cell and/or tissue in vitro and in vivo.9-11 In addition, combination treatment

with TRAIL and chemotherapy showed a synergistic effect in human cancer cell lines

and tumor xenograft models, even in drug-resistant cells and tumors.12-14

As not only probiotics but also anaerobe, Bifidobacterium longum (B. longum) can

be used as a useful host cell for heterologous protein production due to the following

reasons: i) protection of the intestinal mucosa and suppression of the permanent

planting of detrimental gut microorganisms;15,16 ii) induction of the releasing of

endogenous mediator with immunological activity and enhancement of immunity of

host;17 and iii) inhibition of tumor growth by suppressing cancer related genes,

selective localization and proliferation within hypoxic tumors.18-22 However, the fact

that exogenous plasmids can not replicate stably in Bifidobacterium limits the

application of Bifidobacterium in cancer gene therapy. Though some

Bifidobacterium-E. coli shuttle vectors have been constructed in recent years, only

some of them expressed foreign genes successfully.23-33

4

In our previous studies, we developed a shuttle vector (pBV22210) which could

replicate stably and expressed recombinant human endostatin in both E. coli and B.

longum. B. longum transfected with pBV22210-endostatin showed a clear inhibitory

effect on the growth of mouse solid liver tumor in vivo.30 Taking advantage of the

stability of pBV22210 in B. longum, the tumor site-specific localization of B. longum

and the induction of apoptosis in cancer cells by TRAIL, we reported here the cloning

of the extracellular domain of TRAIL (amino acid 114-281) to the downstream of

His-tag of pBV22210, the transfection of B. longum, and the characterization of the

transfectants. Anti-tumor activity of B. longum-pBV22210-TRAIL combining with

chemotherapeutic drugs or B. longum-pBV22210-endostatin in xenograft models of

S180 osteosarcoma was examined.

Materials and Methods

Bacterial strains and plasmids

E. coil strain DH5α, wild-type (WT) B. longum, B. longum-pBV22210-endostatin,

and plasmid pBV22210-endostatin were preserved in our laboratory. pET28a-TRAIL

was a generous gift from Professor Yayi Hou at School of Medicine, Nanjing

University.

Reagents and enzymes

5

Adriamycin (batch C60705) was obtained from Haizheng Pharmaceutical

Enterprise (Taizhou, China), Cyclophosphamide (CTX, batch 06122921) was

purchased from Jiangsu Hengrui Medicine Co, LTD. (Lianyungang, China). 2×Taq

mix was purchased from Tiangen Biotech (Beijing, China), T4 DNA ligase and

restriction endonucleases were purchased from Takara Bio (Dalian, China). Primers

were synthesized by Shanghai Shenergy Biocolor BioScience & Technology

Company (Shanghai, China).

Animals and tumors

Male Kunming mice (19 ± 2 g) were obtained from Qinglongshan Animal Center

(Nanjing, China). The animals were maintained in a pathogen-free environment at a

constant temperature and supplied with laboratory chow and water ad libitum on a

12-hour dark/light cycle. S180 mouse osteosarcoma (mouse with hydroperitoneum)

was purchased from Jiangsu Institute for Cancer Prevention and Cure (Nanjing,

China). S180 cancer cells (1 × 107) were injected subcutaneously in the right flank of

each mouse to establish tumor models.

Construction of pBV22210-TRAIL

(i) PCR amplification of recombinant soluble TRAIL: The plasmid of

6

pET28a-TRAIL was extracted following the instructions and used as template for

PCR. The sequence was as follow: 5’ CCG GAA TTC GTG AGA GAA AGA GGT

CCT CAG 3’(sense) and 5’CGC GGA TCC TTA GCC AAC TAA AAA GGC CCC

GAA AAA ACT G 3’(antisense); 94°C for 30 s, 54°C for 30 s, 72°C for 30 s, 30

cycles. PCR products were analyzed by electrophoresis on 1% agarose gels.

(ii) Restriction endonuclease digestion of pBV22210-endostatin and TRAIL:

pBV22210-endostatin and TRAIL were digested by EcoRI/BamHI and the restricted

pBV22210 and TRAIL fragments were gel-purified before the ligation of the PCR

products and pBV22210 at the EcoRI/BamHI site. The ligation solution was

transferred into competent cells of DH5α and pBV22210-TRAIL was detected on LB

agar plate with 10 μg/ml chloramphenicol.

-------------------------------

Figure 1 near here ------------------------------

Electroporation

Electrocompetent cells of B. longum were prepared according to Rossi et al.34

Bacteria were re-suspended in about 1/100 of the original culture volume of ice-cold

0.5 M sucrose plus 1 mM ammonium citrate (pH 6.0); 1.0 μg of purified

pBV22210-TRAIL plasmid was added in bacteria suspension and incubated at 4°C

for 2-3 hours. The above mixture was added in a pre-cooled sterile Gene Pulser

disposable cuvette (interelectrode distance 0.2 cm; Bio-Rad, Hercules, CA). The

7

pBV22210-TRAIL plasmids were transfected directly into B. longum by

electroporation in a Bio-Rad Gene-Pulser apparatus at 25 µF and 2.5 kV with the

pulse controller set at 200 Ω. Then the mixture was inoculated in TPY culture and

anaerobiocally cultivated overnight at 37°C. Overnight culture was diluted, plated on

TPY agar plates and monoclones were cultivated alternately in TPY culture with and

without 5 μg/ml chloramphenicol for at least 10 generations. Stable B.

longum-pBV22210-TRAIL with a chloramphenicol resistance was selected.

Determination of B. longum transfected with TRAIL by PCR

Based on the previous report,35 a modified method was used to extract plasmids

from B. longum-pBV22210-TRAIL and WT B. longum. Briefly, 10 ml B.

longum-pBV22210-TRAIL cells from overnight TPY culture medium were harvested

and washed twice with PBS before treated with lysozyme (40 mg/ml) at 37°C for 30

min. The plasmids were extracted according to the instructions of 3S Spin Plasmid

Miniprep Kit (Shenergy Biocolor, China). The plasmids were used as template for

PCR under the same protocols stated above.

Carbohydrate fermentation assay

TPY broth supplemented with different carbon sources was used for the

determination of carbohydrate fermentation, with bromocresol purple (0.04 g/l) as a

8

pH indicator.27 After anaerobic incubation in the medium at 37°C for 8 days, B.

longum-pBV22210-endostatin and WT B. longum were detected by the color of

indicator.

Growth assay and morphologic analysis

Overnight culture of B. longum-pBV22210-TRAIL in TPY medium was

inoculated into fresh medium with and without 5 μg/ml chloramphenicol to an initial

optical density value of 0.010 at 600 nm (OD 600). The cultures were grown under

anaerobic conditions at 37°C for 24 hr and OD values were measured every hour. WT

B. longum cells were used as a control.

B. longum-pBV22210-TRAIL and WT B. longum cells were cultured in TPY

medium till they grew to a logarithmic phase, harvested and washed twice with 0.1 M

PBS (pH 7.0 ) by centrifugation. After smeared and affixed on a slide, the cells were

stained with Gram stain reagent following the instructions and examined under ×

1000 objective of a light microscope.

Examination of localization of B. longum-pBV22210-TRAIL in mice

72 hours after hypodermic inoculation of S180 cells, the tumor-bearing mice were

injected with the same dose of B. longum-pBV22210-TRAIL (1×108 cells/ mouse) via

9

tail vein every 24 hours for three times. At 1, 24, 48 and 96 hr after the third injection,

the tumor-bearing mice were sacrificed (five per time), and tissue samples were

obtained from the heart, liver, spleen, lung, kidney and tumor under aseptic conditions.

Each tissue sample (0.1 g) was homogenized in 1 ml dextrose-saline solution (5%

glucose in 0.9% NaCl). The homogenates were plated on the solid TPY medium

(1.5% agar) and incubated anaerobically at 37°C for 48 hours. The number of

colonies per dish was counted to determine the number of viable bacteria. Other

tumor-bearing mice were injected with ampicillin (50 mg/kg) at 96 hours after the

third injection of B. longum-pBV22210-TRAIL, the tumors were excised from mice

at 24, 48 and 96 hr after the injection of ampicillin. The homogenates of tumors were

treated and incubated on the solid TPY medium (1.5% agar) as above.

Antitumor activity of B. longum-pBV22210-TRAIL and B. longum-pBV22210-

endostatin

The mice were weighted and divided into six groups randomly after hypodermic

inoculation of S180 cells for 24 hr. Mice as negative control were injected with

dextrose-saline solution (0.4 ml/day, i.v, on days 1-7). Five treated groups were

injected respectively with B. longum-pBV22210-TRAIL (0.4 ml/day, i.v, on days 1-7),

B. longum-pBV22210-endostatin (0.4 ml/day, i.v, on days 1-7), WT B. longum (0.4

ml/day, i.v, on days 1-7), combination of B. longum-pBV22210-TRAIL and B.

longum-pBV22210-endostatin (0.2 ml + 0.2 ml/day, i.v, on days 1-7), CTX (30 mg/kg,

10

i.p, on days 4, 6, 8). All B. longum cells were washed and re-suspended in

dextrose-saline solution at a concentration of 2.5×108 cells/ml before injection. CTX

was suspended in dextrose-saline solution and 0.2 ml of the solution was injected each

time per mice. The mice were sacrificed on day 10, the tumors were excised and

weighed. The volume of the tumors was calculated using the formula: tumor volume

= (width) 2 × length × 0.52.36 The inhibition of tumor growth was determined by the

following formula:

tumor weight/volume of control group - tumor weight/volume of treatment group

tumor weight/volume of control group ×100%

Antitumor activity of B. longum-pBV22210-TRAIL combining chemotherapeutic

drugs

The mice were weighed and randomized into four groups (six per group) after

hypodermic inoculation of S180 cells for 24hr. Mice as negative control were injected

with dextrose-saline solution containing 5% glucose and 0.9% NaCl (0.4 ml/day, i.v,

on days 1-7), and the mice in the three treated groups were injected with B.

longum-pBV22210-TRAIL (0.4 ml/day, i.v, on days 1-7), Adriamycin (5 mg/kg, i.p,

on day 2), combination of B. longum-pBV22210-TRAIL (0.4 ml/day, i.v, on days 1-7)

and Adriamycin (5 mg/kg, i.p, on day 2), respectively. All B. longum cells were

washed three times and re-suspended before injection as mentioned above. After

injection intravenously via the tail vein on day 7, the mice were kept for an additional

3 days till they were sacrificed on day 10. The weight and volume of the excised

tumors and the inhibition of tumor growth were determined as described above.

11

Statistical Analysis

The data were statistically analyzed using Student’s t-test in both groups and

ANOVA test in multiple groups. The comparisons among multiple groups were

performed using Student–Newman–Keuls q-test. P values of < 0.05 were considered

to be significant.

Results

Identification of B. longum-pBV22210-TRAIL

The plasmids extracted from B. longum transfected with pBV-22210-TRAIL were

amplified by PCR and the results were shown in Fig. 2. A band of about 500 bp was

detected in B. longum-pBV22210-TRAIL but not in WT B. longum that served as a

negative control.

-------------------------------

Figure 2 near here ------------------------------

Carbohydrate fermentation

The results demonstrated that the characteristics of B. longum-pBV22210-TRAIL

were concordant to that of WT B. longum cells in carbohydrate fermentation except

12

for three carbohydrates: salicin, mannose and melezitose (Table 1). B.

longum-pBV22210-TRAIL cells fermented salicin and mannose but not melezitose

while WT B. longum cells showed contrary results.

------------------------------

Table 1 near here ------------------------------

Growth and morphological characteristics

The growth curve of B. longum-pBV22210-TRAIL and WT B. longum cells were

similar in TPY medium without selective pressure while the lag phase of B.

longum-pBV22210-TRAIL in selective medium was statistically longer than that in

nonselective medium as shown in Fig. 3. Both B. longum-pBV22210-TRAIL and WT

B. longum cells in TPY medium without chloramphenicol grew to an exponential

phase after 6 hr and stationary phase (OD 600=1.1) after 10 hr of incubation.

However, B. longum-pBV22210-TRAIL cells in TPY medium with chloramphenicol

(5 μg/ml) grew to an exponential phase after 15 hr and stationary phase after 18 hr of

incubation.

------------------------------

Figure 3 near here ------------------------------

Localization and decolonization of B. longum-pBV22210-TRAIL in tumor tissue

72 hours after subcutaneous inoculation with S180 cells, tumor-bearing mice were

intravenously injected with B. longum-pBV22210-TRAIL cells. Tumor-bearing mice

13

were sacrificed at 1, 24, 48 and 96 hours after the third injection of B. longum-

pBV22210-TRAIL cells (five mice per time) and the localization of transformed B.

longum in tumors and several normal tissues were detected. Other mice were

sacrificed at 24, 48, 96 hours after an additional injection of ampicillin at 96 hr

following the treatment of B. longum-pBV22210-TRAIL cells and the presence of

transformed B. longum in tumors was also detected. At 96 hours, about 1.55 × 107

bacilli/g tumor tissue was found, but few bacilli were detected in normal tissues such

as the heart, liver, spleen, lung, kidney from the tumor-bearing mice (Fig. 4). The

increasing number of bacilli in tumors suggested that the transformed B. lignum

selectively proliferated in the tumor tissue. In contrast, the number of transformed B.

longum in normal tissues decreased rapidly 24hr after injection indicated that

transformed B. longum did not germinate in normal tissues. Interestingly, the number

of transformed B. longum located in tumors decreased rapidly after the treatment of

ampicillin, few viable bacilli was detected in tumors 96 hours after the administration

of ampicillin.

------------------------------

Figure 4 near here ------------------------------

Effect of B. longum-pBV22210-TRAIL and B. longum-pBV22210-endostatin on

growth of S180 tumor xenografts

We established a xenograft model to assess the efficacy of B. longum

-pBV22210-TRAIL and B. longum-pBV22210-endostatin. The tumors excised from

14

each group were shown in Fig. 5A. Compared to dextrose-saline solution-treated

group, treatment with B. longum-pBV22210-TRAIL combining B.

longum-pBV22210-endostatin significantly suppressed tumor weight by 79.6% (p =

0.0034) and volume by 82.6 % (p = 0.001), comparing with B.

longum-pBV22210-TRAIL alone by 58.0% (p = 0.006) and by 60.9% (p = 0.0038), B.

longum-pBV22210-Endostatin alone by 55.3% (p = 0.0074) and by 58.1% (p =

0.0037), and CTX by 63.4% (p = 0.0051) and by 65.1% (p = 0.0026), respectively;

Compared with WT B. longum, combination treatment inhibited the tumor growth as

measured by tumor weight by 73.6% (p = 0.0077) and by tumor volume by 76.7% (p

= 0.0006) while B. longum-pBV22210-TRAIL alone respectively by 45.6% (p =

0.0229) and by 47.7% (p = 0.0059) and B. longum-pBV22210-endostatin alone by

42.1% (p = 0.0299) and by 40.0% (p = 0.0074), respectively. The inhibitory effect of

combination treatment was 28.0 % (p = 0.054) higher as measured by tumor weight

and 29% (p = 0.001) higher by tumor volume than those of B.

longum-pBV22210-TRAIL treatment alone; and it was 31.5% (p = 0.033) and 36.7%

(p = 0.003) higher as measured by tumor weight and by volume than those of B.

longum-pBV22210-endostatin treatment alone, respectively. Based on these results,

we could conclude that the combination of B. longum-pBV22210-TRAIL and B.

longum-pBV22210-endostatin exhibited synergistic effect on tumor inhibition.

------------------------------

Figure 5 near here ------------------------------

Effect of B. longum-pBV22210-TRAIL and Adriamycin on growth of S180 tumor

15

xenografts

To examine the activity of B. longum-pBV22210-TRAIL cells in vivo, we

developed a xenograft model in which S180 cells were injected s.c. into Kunming

mice. Tumors grew rapidly in dextrose-saline solution-treated mice and slower in

treated groups, especially in the combination treatment group. The tumors excised

from each group were shown in Fig. 6A. Compared to dextrose-saline solution group,

combination treatment with B. longum-pBV22210-TRAIL and low dose Adriamycin

markedly inhibited tumor growth by 70.0% (p = 0.0009) as measured by weight and

by 75.5% (p = 0.0002) as measured by volume; B. longum-pBV22210-TRAIL alone

inhibited tumor growth by 56.0% (p = 0.0026) by weight and by 59.2% (p = 0.0013)

by volume; low dose Adriamycin alone by 39.5 % (p = 0.0222) by weight and by

34.2% (p = 0.0671) by volume, respectively (Fig. 6B and C). Combination treatment

enhanced tumor inhibition rate by 14.0% (p = 0.052) and 20.5% (p = 0.027) as

measured by tumor weight and by 16.3% (p= 0.099) and 41.3% (p = 0.033) as

measured by tumor volume, respectively, compared to either agents used alone.

Therefore, combination of B. longum-pBV22210-TRAIL and Adriamycin showed a

more effective inhibition in tumor growth than either B. longum-pBV22210-TRAIL

alone or low dose Adriamycin alone.

-------------------------------

Figure 6 near here ------------------------------

Discussion

16

Although the advantages of Bifidobacterium as a gene delivery vehicle have been

confirmed for several decades, the instability of exogenous plasmid and low-level

expression of exogenous gene hamper its further application in cancer gene therapy.

Certain progress has been made in the field over the years. Yi et al (2005)

successfully constructed a Bifidobacterium infantis/CD targeting gene therapy system

with a recombinant CD/pGEX-1LamdaT plasmid.37 Sasaki et al (2002) transfected

WT B. longum with pBLES100-S-eCD to produce CD in hypoxic tumors and

achieved tumor site-specific conversion of 5-FC to 5-FU that resulted in significant

antitumor effect in rat bearing autochthonous mammary tumors not only by

intratumoral injection but also by systemic administration of transfected B. longum.26

Fu et al (2005) used transformed B. longum carrying shuttle vector pBV220 (Amp+)

encoding human endostatin as a delivery system and succeeded in selective

localization within solid tumors and inhibiting growth of liver tumor xenografts in

mice.28 However, ampicillin is considered to be harmful to bacterial cytoderm

synthesis after electroporation, Xu et al (2007) constructed a new vector

pBV22210-endostatin combining a chloramphenicol resistance gene and a cryptic

plasmid pMB1 from WT B. longum strain that made the transformed B. longum more

stable, and the expressed recombinant protein from B. longum-pBV22210-endostatin

exhibited stronger suppression of tumor growth in xenografts models than that from B.

longum-pBV220-endostatin in a previous study.30 Based on the previous work, we

further constructed a new plasmid B. longum-pBV22210-TRAIL that encoded the

17

extracellular domain of TRAIL (Fig. 2) and expressed recombinant human TRAIL in

B. longum successfully (data not shown) in the present study.

Maintenance of the inborn characteristics of B. longum is of great significance for

retaining the physiological role of B. longum. Carbohydrate fermentation indicates the

ability of transformed B. longum to use carbon source and it is considered to be an

important basis to identify bacterium strain. The results showed that B.

longum-pBV22210-TRAIL differed from WT B. longum cells in carbohydrate

fermentation in salicin, mannose and melezitose. The characteristics of two type B.

longum in carbohydrate fermentation were essentially consistent as it was reported

that the fermentation of L-arabinose, gluconate, inulin, lactose, D-mannose, methyl

a-D-glucoside, ribose, salicin, trehalose or xylose was variable in B. longum.38

Morphology and growth curve were two additional indexes that illustrate the main

biochemical characteristics of transformed B. longum. Therefore, we further

determined the morphology and growth curve to confirm that transfection of TRAIL

did not alter the main biochemical characteristics. Our results indicated that B.

longum-pBV22210-TRAIL exhibited similar properties in biochemical, growth and

morphological characteristics (data not shown) as WT B. longum, and it was

consistent with Xu et al’s study.30

WT B. longum and transformed B. longum could both specifically reach tumor

tissue by circulation and grow in hypoxic solid tumors as reported by Yazawa K et al

18

(2001).25 Our results were consistent with the observation by demonstrating the

presence of the viable bacilli of transformed B. longum in several main organs of

treated mice. The relatively hypoxia region and abundance of nutrition in tumor offer

the possibility of B. longum cells’ colonization. However, the mechanism of B.

longum’ localization to the tumor tissue from the blood has not been illustrated. It is

possible that thin vessel walls and wide intercellular space between endothelial cells

in the special vascular structure of solid tumors contribute to high permeability of

blood vessels which may make the translocation of B. longum easier and quicker.

Since Bifidobacterium can be killed easily by antibiotics such as kanamycin,

cefoperazone, and penicillin in vitro as shown in our previous study30, we tested B.

longum’ sensitivity to ampicillin in a decolonization assay and found that the number

of B. longum located in tumors decreased drastically after an injection of ampicillin

following the treatment of B. longum-pBV22210-TRAIL. The result implicated that

the expression of target gene could be controlled, and it is helpful for further

application of B. longum.

The suppressive effect of B. longum-pBV22210-TRAIL on tumor growth was also

determined in osteosarcoma xenograft models in vivo. Since B.

longum-pBV22210-endostatin was proved to have definite inhibitive effect on tumor

growth in our previous study,30 we examined the activity of combination of B.

longum-pBV22210-TRAIL and B. longum-pBV22210-endostatin and assessed the

synergistic effect of the combination treatment. We propose two possible mechanisms

19

for the phenomenon: on the one hand, endostatin as an angiogenesis inhibitor

depressed the vascularization and reduced provision of nutrition in tumor growth; on

the other hand, TRAIL induced apoptosis of tumor cells through binding to its

receptors and activating related apoptosis pathways. Thus, tumor growth was

significantly suppressed by the combination of TRAIL and endostatin. The tumor

inhibition rate in combination group was 20% higher than that in B.

longum-pBV22210-TRAIL or B. longum-pBV22210-endostatin group when used

alone. Furthermore, since many chemotherapeutic drugs have severe side effects, we

chose low dose Adriamycin to combine with TRAIL and determined their synergistic

effect in tumor growth inhibition. Combination of B. longum-pBV22210-TRAIL and

low dose Adriamycin (5 mg/kg) resulted in the higher inhibition of tumor growth.

Similarly, the sequential treatment with chemotherapeutic drugs followed by TRAIL

induced apoptosis in breast tumor cells and resulted in the inhibition of tumor growth

and the improved survival rate of tumor-bearing nude mice.39 Jin et al (2004)

demonstrated that administration of TRAIL plus Taxol and Carboplatin led to

suppression of tumor growth and improved survival significantly in both

subcutaneous and orthotopic lung tumor xenograft models.40 The mechanism for

chemotherapeutics enhancing TRAIL-induced apoptosis in tumor cells has not been

completely elucidated. A recent study showed that treatment with subtoxic doses of

silibinin in combination with TRAIL induced rapid apoptosis in TRAIL-resistant

glioma cells through up-regulating DR5 and down-regulating the levels of the

antiapoptotic proteins FLIPL, FLIPS, and survivin.41 After pretreated with

20

cis-Diaminedichloroplatinum (CDDP), Etoposide (VP-16), Adriamycin, and

vincristine, sensitization of TRAIL- and drug-resistant prostate carcinoma PC-3 cells

to TRAIL-induced apoptosis was enhanced via inhibition of the transcription

repressor Yin Yang 1 (YY1) and up-regulation of DR5 expression.42 Other reports also

indicated that antineoplastic agents were capable of up-regulating levels of DRs (DR4

and DR5),43-45 inducing (e.g., Bax and Bak) or reducing (e.g., Bc1-2 and Bcl-xL)

proapoptotic Bcl-2 family members,46,47 changing the relative levels of RelA and

c-Rel subunits of NF-κB.48 Moreover, it has been shown that the synergistic effects of

chemotherapeutic drugs and TRAIL on apoptosis occur through activation of

downstream caspase-3, which can be activated by both mitochondria-dependent and

-independent pathways.39 Therefore, chemotherapy can sensitize tumor cells to

TRAIL-induced apoptosis in part through death receptors up-regulation, and in part

through cross-talk between the intrinsic and extrinsic pathways. Combination of

TRAIL with chemotherapy resulted in reversal of resistance to TRAIL-mediated

apoptosis and enhancement of TRAIL’s efficiency, however, sensitivity of human

normal cells to such combinations is not well known. A recent study showed that

TRAIL/cisplatin showed toxicity towards human primary hepatocytes and resting

lymphocytes, both TRAIL/5-fluorouracil and TRAIL/cisplatin combinations are toxic

toward PHA-IL2-activated lymphocytes.49 These results suggest the importance to

conduct cytotoxicity study of TRAIL/anticancer drug combinations in normal cells.

In summary, we have developed a strategy of combining TRAIL with

21

antineoplastic agents or other genes (e.g., endostatin) for the treatment of

osteosarcoma by using B. longum as an effective delivery system. The combination

treatment resulted in significant inhibition of tumor growth, comparing with TRAIL,

chemotherapy or endostatin treatment alone. Our results support the potential of

combination of chemotherapy and tumor therapeutic target genes to provide a novel,

selective localization and apoptosis-based biological approach for advancing the

treatment of cancer.

22

Acknowledgements

This work was supported by grant 2006AA02Z19E of the 863 Project from the

State Ministry of Science and Technology of China, the 985-II Project from Nanjing

University and grant BK2008150 from the Natural Science Foundation of Jiangsu

Province to GXX; and grant 30670671 from the National Natural Science Foundation

of China, grant BK2006713 from the Natural Science Foundation of Jiangsu Province,

China and RFDP grant 20050284025 from the State Educational Ministry of China to

JJW.

23

References

1 Pitti RM, Marsters SA, Ruppert S, Donahue CJ, Moore A, Aahkenazi A.

Induction of apoptosis by Apo-2 ligand, a new member of the tumor necrosis

factor family. J Biol Chem 1996; 271: 12687-12690.

2 Wiley SR, Schooley K, Smoolak PJ, Din WS, Huang CP, Nicholl JK et al.

Identification and characterization of a new member of the TNF family that

induces apoptosis. Immunity 1995; 3: 673-682.

3 Sheridan JP, Marsters SA, Pitti RM, Gurney A, Skubatch M, Baldwin D et al.

Control of TRAIL-induced apoptosis by a family of signaling and decoy

receptors. Science 1997; 277: 818-821.

4 Wu GS, Burns TF, McDonald ER 3rd, Jiang W, Meng R, Krantz ID et al.

KILLER/DR5 is a DNA damage-inducible p53-regulated death receptor gene.

Nat Genet 1997; 17: 141-143.

5 Chaudhary PM, Eby M, Jasmin A, Bookwalter A, Murray J, Hood L. Death

receptor 5, a new member of the TNFR family, and DR4 induce

FADD-dependent apoptosis and activate the NF-kB pathway. Immunity 1997; 7:

821-830.

6 Pan G, Ni J, Wei YF Yu G, Gentz R, Dixit VM. An antagonist decoy receptor

and a new death domain-containing receptors for TRAIL. Science 1997; 277:

815-818.

7 Screaton GR, Mongkolsapaya J, Xu XN, Cowper AE, McMichael AJ, Bell JI.

24

TRICK2, a new alternatively spliced receptor that transduces the cytotoxic

signal from TRAIL. Curr Biol 1997; 7: 693-696.

8 Ashkenazi A. Targeting death and decoy receptors of the tumour necrosis factor

superfamily. Nature Rev Cancer 2002; 2: 420-430.

9 Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA, Marsters SA et al.

Safety and antitumor activity of recombinant soluble Apo2 ligand. J Clin

Investig 1999; 104: 155-162.

10 Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M et al.

Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing ligand

in vivo. Nat Med 1999; 5: 157-163.

11 Roth W, Isenmann S, Naumann U, Kügler S, Bähr M, Dichgans J et al.

Locoregional Apo2L/TRAIL eradicates intracranial human malignant glioma

xenografts in athymic mice in the absence of neurotoxicity. Biochem Biophys

Res Commun 1999; 265: 479-483.

12 Gliniak B, Le T. Tumor necrosis factor-related apoptosis-inducing ligand’s

antitumor activity in vivo is enhanced by the chemotherapeutic agent CPT-11.

Cancer Res 1999; 59: 6153-6158.

13 Nimmanapalli R., Porosnicu M, Nguyen D, Worthington E, O’Bryan E, Perkins

C et al. Cotreatment with STI-571 enhances tumor necrosis factor-related

apoptosis-inducing ligand (TRAIL or apo-2L)-induced apoptosis of

Bcr-Abl-positive human acute leukemia cells. Clin Cancer Res 2001; 7:

350-357.

25

14 Naka T, Sugamura K, Hylander BL, Widmer MB, Rustum YM, Repasky EA.

Effects of tumor necrosis factor-related apoptosis-inducing ligand alone and in

combination with chemotherapeutic agents on patients’ colon tumors grown in

SCID mice. Cancer Res 2002; 62: 5800-5806.

15 Marteau PR, de Vrese M, Cellier CJ, Shrezenmeir J. Protection from

gastrointestinal diseases with the use of probiotics. Am J Clin Nutr 2001; 73:

430S-436S.

16 Ito M, Sawada H, Ohishi K, Yoshida Y, Yokoi W, Watanabe T et al.

Suppressive effects of Bifidobacteria on lipid peroxidation in the colonic

mucosa of iron-overloaded mice. J Dairy Sci 2001; 84:1583-1589.

17 Yasui H, Ohwaki M. Enhancement of immune response in Peyer’s patch cells

cultured with Bifidobacterium breve. J Dairy Sci 1991; 74:1187-1195.

18 Kimura NT, Taniguchi S, Aoki K, Baba T. Selective localization and growth of

Bifidobacterium bifidum in mouse tumors following intravenous administration.

Cancer Res 1980; 40: 2061-2068.

19 Yazawa K, Fujimori M, Amano J, Kano Y, Taniguchi S. Bifidobacterium

longum as a delivery system for cancer gene therapy: Selective location and

growth in hypoxic tumors. Cancer Gene Ther 2000; 7: 269-274.

20 Singh J, Rivenson A, Tomita M, Shimamura S, Ishibashi N, Reddy BS.

Bifidobacterium longum, a lactic acid-producing intestinal bacterium inhibits

colon cancer and modulates the intermediate biomarkers of colon

carcinogenesis. Carcinogenesis 1997; 18: 833-841.

26

21 Fujimori M, Amano J, Taniguchi S. The genus Bifidobacterium for cancer gene

therapy. Curr Opin Drug Discov Devel 2002, 5: 200-203.

22 Pawelek J, Low K, Bermudes D. Bacteria as tumour-targeting vectors. Lancet

Oncol 2003; 4: 548-556.

23 Rossi M, Brigidi P, Matteuzzi D. Improved cloning vectors for Bifidobacterium

spp. Lett Appl Microbiol 1998; 26:101-104.

24 Rossi M, Brigidi P, Gonzalez Vara y Rodriguez A, Matteuzzi D.

Characterization of the plasmid pMB1 from Bifidobacterium longum and its use

for shuttle vector construction. Res Microbiol 1996; 147: 133-143.

25 Yazawa K, Fujimori M, Nakamura T, Sasaki T, Amano J, Kano Y et al.

Bifidobacterium longum as a delivery system for gene therapy of chemically

induced rat mammary tumors. Breast Cancer Res Treat 2001; 66: 165-170.

26 Nakamura T, Sasaki T, Fujimori M, Yazawa K, Kano Y, Amano J et al. Cloned

cytosine deaminase gene expression of Bifidobacterium longum and application

to enzyme/pro-drug therapy of hypoxic solid tumors. Biosci Biotechnol

Biochem 2002; 66: 2362-2366.

27 Li X, Fu GF, Fan YR, Liu WH, Liu XJ, Wang JJ et al. Bifidobacterium

adolescentis as a delivery system of endostatin for cancer gene therapy:

selective inhibitor of angiogenesis and hypoxic tumor growth. Cancer Gene

Ther 2003; 10:105-111.

28 Fu GF, Li X, Hou YY, Fan YR, Liu WH, Xu GX. Bifidobacterium longum as an

oral delivery system of endostatin for gene therapy on solid liver cancer. Cancer

27

Gene Ther 2005; 12:133-140.

29 Yi C, Huang Y, Guo ZY, Wang SR. Antitumor effect of cytosine

deaminase/5-fluorocytosine suicide gene therapy system mediated by

Bifidobacterium infantis on melanoma. Acta Pharmacol Sin 2005; 26: 629-634.

30 Xu YF, Zhu LP, Hu B, Fu GF, Zhang HY, Wang JJ, Xu GX. A new expressing

plasmid in Bifidobacterium Longum as a delivery system of endostatin for

cancer gene therapy. Cancer Gene Ther 2007; 14: 151-157.

31 Park MS, Kwon B, Shim JJ, Huh CS, Ji GE. Heterologous expression of

cholesterol oxidase in Bifidobacterium longum under the control of 16S rRNA

gene promoter of bifidobacteria. Biotechnol Lett 2008; 30: 165-172.

32 Reyes Escogido ML, De Leon Rodriguez A, Barba de la Rosa AP. A novel

binary expression vector for production of human IL-10 in Escherichia coli and

Bifidobacterium longum. Biotechnol Lett 2007; 29: 1249-1253.

33 Guglielmetti S, Ciranna A, Mora D, Parini C, Karp M. Construction,

characterization and exemplificative application of bioluminescent

Bifidobacterium longum biovar longum. Int J Food Microbiol 2008; 124:

285-290.

34 Rossi M, Brigidi P, Matteuzzi D. An efficient transformation system for

Bifidobacterium spp. Lett Appl Microbiol 1997; 24: 33-36.

35 Park MS, Lee KH, Ji GE. Isolation and characterization of two plasmids from

Bifidobacterium longum. Lett Appl Microbiol 1997; 25: 5-7.

36 Boehm T, Folkman J, Browder T, O'Reilly MS. Antiangiogenic therapy of

28

experimental cancer does not induce acquired drug resistance. Nature 1997; 390:

404-407.

37 Yi C, Huang Y, Guo Z, Wang S. Construction of Bifidobacterium Infantis/CD

targeting gene therapy system. The Chinese-German Journal of Clinical

Oncology 2005; 4: 244-247.

38 Zhu L, Li W, Dong X. Species identification of genus Bifidobacterium based on

partial HSP60 gene sequences and proposal of Bifidobacterium

thermacidophilum subsp. Porcinum subsp. nov. Int J Syst Evol Microbiol 2003;

53: 1619-1623

39 Singh TR, Shankar S, Chen X, Asim M, Srivastava RK. Synergistic interactions

of chemotherapeutic drugs and tumor necrosis factor related apoptosis-inducing

ligand/Apo-2 ligand on apoptosis and on inhibition of breast carcinoma in vivo.

Cancer Res 2003; 63: 5390-5400.

40 Jin H, Yang R, Fong S, Totpal K, Lawrence D, Zheng Z et al. Apo2 ligand/

tumor necrosis factor related apoptosis-inducing ligand cooperates with

chemotherapy to inhibit orthotopic lung tumor growth and improve survival.

Cancer Res 2004; 64: 4900-4905.

41 Son YG, Kim EH, Kim JY, Kim SU, Kwon TK, Yoon AR et al. Silibinin

Sensitizes Human Glioma Cells to TRAIL-Mediated Apoptosis via DR5

Up-regulation and Down-regulation of c-FLIP and Survivin. Cancer Res 2007;

67: 8274-8284.

42 Baritaki S, Huerta-Yepez S, Sakai T, Spandidos DA, Bonavida B.

29

Chemotherapeutic drugs sensitize cancer cells to TRAIL-mediated apoptosis:

up-regulation of DR5 and inhibition of Yin Yang 1. Mol Cancer Ther 2007; 6:

1387-1399.

43 Wen J, Ramadevi N, Nguyen D, Perkins C, Worthington E, Bhalla K.

Antileukemic drugs increase death receptor 5 levels and enhance Apo-2L-

induced apoptosis of human acute leukemia cells. Blood 2000; 96: 3900-3906.

44 Gibson SB, Oyer R, Spalding AC, Anderson SM, Johnson GL. Increased

expression of death receptors 4 and 5 synergizes the apoptosis response to

combined treatment with etoposide and TRAIL. Mol Cell Biol 2000; 20:

205-212.

45 Matsui T, Sowa Y, Yoshida T, Murata H, Horinaka M, Wakada M et al.

Sulforaphane enhances TRAIL-induced apoptosis through the induction of DR5

expression in human osteosarcoma cells. Carcinogenesis 2006; 27: 1768-1777.

46 Villunger A, Michalak EM, Coultas L, Müllauer F, Böck G, Ausserlechner MJ et

al. p53- and drug-induced apoptotic responses mediated by BH3-only proteins

puma and noxa. Science 2003; 302: 1036-1038.

47 Siervo-Sassi RR, Marrangoni AM, Feng X, Naoumova N, Winans M, Edwards

RP et al. Physiological and molecular effects of Apo2L/TRAIL and cisplatin in

ovarian carcinoma cell lines.Cancer Lett. 2003; 190: 61-72.

48 Chen X, Kandasamy K, Srivastava RK. Differential roles of RelA (p65) and

c-Rel subunits of nuclear factor κB in tumor necrosis factor-related apoptosis

inducing ligand signaling. Cancer Res 2003; 63: 1059-1066.

30

49 Meurette O, Fontaine A, Rebillard A, Le Moigne G, Lamy T, Lagadic-gossmann

D et al. Cytotoxicity of TRAIL/anticancer drug combinations in human normal

cells. Ann. N.Y. Acad. Sci. 2006; 1090: 209-216.

31

Titles and legends to figures

Figure 1. Construction of the expression vector pBV22210-TRAIL.

Figure 2. Identification of TRAIL gene in transformed B. longum by PCR analysis.

Lane 1: DNA maker (2000, 1500, 1000, 750, 500, 250, 100); Lane 2: WT B. longum

as negative control; Lane 3: B. longum transformed with pBV22210-TRAIL plasmid.

Figure 3. The growth curves of B. longum-pBV22210-TRAIL and WT B. longum

cells. B. longum-pBV22210-TRAIL cells were incubated anaerobically at 37°C in

TPY medium with 5 μg/ml chloramphenicol (filled squares) or without

chloramphenicol (filled triangles). WT B. longum were incubated anaerobically in

TPY medium without selective pressure (open squares). OD 600, optical density at

600 nm.

Figure 4. Viable bacilli number of B. longum-pBV22210-TRAIL in tumors and

normal organs of treated mice. (A) shows the distribution of B. longum-pBV22210-

TRAIL in different normal organs at different time after the third injection of 1 × 108

viable bacilli. (B) shows the viable bacilli number in tumors after administration of

ampicillin (50 mg/kg) following the treatment of B. longum-pBV22210-TRAIL cells.

Figure 5. The inhibition effects on S180 tumor growth by B. longum-pBV22210-

32

TRAIL cells and/or B. longum-pBV22210-endostatin cells in tumor-bearing mice.

There were five mice in each group. The tumor weights and tumor volumes were

measured for each mouse. (A) shows the tumors excised from tumor-bearing mice.

Row 1, dextrose-saline solution group; row 2, WT B. longum cells group; row 3, B.

longum-pBV22210-endostatin cells group; row 4, B. longum-pBV22210-TRAIL cells

group; row 5, B. longum-pBV22210-TRAIL cells combined with B. longum-

pBV22210-endostatin cells group; row 6, CTX group. (B) and (C) show the average

tumor weight and average tumor volume, respectively. Groups in Bars 1-6

corresponded to that in Row 1-6. Both tumor weights and tumor volumes were

significantly reduced in B. longum-pBV22210-TRAIL cells combined with B.

longum-pBV22210-endostatin cells group.

Figure 6. The suppression effects on S180 tumor growth of B. longum

-pBV22210-TRAIL cells and/or low dose Adriamycin (5 mg/kg) in tumor-bearing

mice. There were six mice in each group. The tumor weights and tumor volumes were

measured for each mouse. (A) shows the tumors excised form tumor-bearing mice.

Rows 1-4 were dextrose-saline solution group, low dose Adriamycin (5 mg/kg) group,

B. longum–pBV22210-TRAIL cells group, B. longum-pBV22210-TRAIL cells plus

Adriamycin (5 mg/kg)group respectively. (B) and (C) show the average tumor weight

and average tumor volume. Groups in Bars 1-4 corresponded to that in Rows 1-4.

Both tumor weights and tumor volumes were significantly reduced in drug

combination treated mice. * p < 0.05, ** p < 0.01.

33

Table

Table 1 The carbohydrates fermentation of B. longum-pBV22210-TRAIL cells (A)

and WT B. longum cells (B)

Ara Cel Fru Gal Glu Inu Lac Mal Man Mel Raf Rib Sal Sor Sta Suc Xyl

A + − + + + − + + + − + + + − − + +

B + − + + + − + + − + + + − − − + +

Carbon source: (Ara): arabinose; (Cel): cellobiose; (Fru): fructose; (Gal): galactose;

(Glu): glucose; (Inu): inulin; (Lac): lactose; (Mal): maltose; (Man): mannose; (Mel):

melezitose; (Raf): raffinose; (Rib): ribose; (Sal): salicin; (Sor): sorbitol; (Sta): starch;

(Suc): sucrose; (Xyl): xylose. (+),: positive reaction; (−),: negative reaction.